Swarm’s inclusion in the official go-ethereum release as an experimental feature was eagerly anticipated with the release of geth 1.5, aptly named “let there bee light.” The current version of the code, POC 0.2 RC5, bears the name “embrace your daemons” and represents a refactored and cleaner iteration of the codebase that has been utilized on the Swarm toynet in recent months.
In this latest release, a standalone Swarm daemon can be launched through the swarmcommand, which operates as a separate process and can be integrated with your preferred IPC-compliant ethereum client if necessary. The smooth operation and efficient delivery of content are facilitated by bandwidth accounting, employing the Swarm Accounting Protocol (SWAP). This protocol incentivizes nodes to contribute their bandwidth and relay data. While the SWAP system is functional, it is disabled by default. Additionally, POC 0.4 aims to introduce storage incentives (punitive insurance) to ensure the availability of rarely-accessed content. Presently, the client primarily utilizes the blockchain for domain name resolution as the default setting.
In summary, the latest release of Swarm brings its experimental features to the official go-ethereum release, offering enhanced functionality, improved codebase, and the potential for incentivized bandwidth contribution and efficient content delivery.
We are thrilled to announce the launch of our new Swarm testnet, which is now connected to the Ropsten ethereum testchain. The Ethereum Foundation has generously contributed a Swarm cluster consisting of 35 nodes (expected to expand to 105) running on the Azure cloud. This cluster is hosting the Swarm homepage.
This testnet represents our first public pilot, and we warmly invite the community to join the network, contribute resources, report issues, identify pain points, and provide feedback on usability. Detailed instructions can be found in the Swarm guide. We especially encourage individuals who are capable of running persistent nodes (nodes that remain online) to reach out to us. We have already received commitments for 100TB deployments. However, please note that the testnet does not offer any guarantees. Data loss or unavailability may occur. It is important to emphasize that guarantees of persistence cannot be assured until the storage insurance incentive layer is implemented.
We want to reiterate that the testnet comes with no guarantees, and there is a possibility of data loss or unavailability. We are committed to working towards implementing the storage insurance incentive layer to ensure greater persistence. Until then, we appreciate the community’s involvement and understanding as we continue to refine and improve the Swarm network.
We envision shaping this project with more and more community involvement, so we are inviting those interested to join
How does Swarm work?
Swarm operates as a distributed storage platform and content distribution service, serving as a foundational component within the Ethereum Web3 stack. Its primary goal is to provide a peer-to-peer storage and serving solution that offers continuous availability, resistance to distributed denial-of-service (DDOS) attacks, fault tolerance, censorship resistance, and self-sustainability through an integrated incentive system.
The incentive layer of Swarm utilizes peer-to-peer accounting for bandwidth allocation, deposit-based storage incentives, and enables the trading of resources in exchange for payment. This ensures that participants are incentivized to contribute their resources to the network and maintain its operation. By implementing these incentives, Swarm aims to create a robust and reliable storage and content distribution infrastructure.
Swarm is designed to deeply integrate with the devp2p multiprotocol network layer of Ethereum, enabling seamless communication and data exchange between nodes. Additionally, it leverages the Ethereum blockchain for domain name resolution, ensuring a cohesive integration with the existing Ethereum ecosystem.
In summary, Swarm functions as a decentralized and secure storage and content distribution platform, utilizing incentives, peer-to-peer networking, and integration with the Ethereum blockchain to provide a reliable and resilient infrastructure for storing and serving data.
Swarm functions as a distributed chunkstore, allowing users to upload content onto the network. The nodes within the Swarm ecosystem contribute their resources such as diskspace, memory, bandwidth, and CPU to store and serve these chunks of data. But how are chunks assigned to specific nodes?
Each Swarm node has an address, which is derived from the hash of their bzz-account address. This address space is referred to as the overlay network. When a chunk is uploaded to Swarm, the protocol determines that it will ultimately be stored at nodes that are closest to the chunk’s address, based on a well-defined distance measure within the overlay address space. This process, known as syncing, is an integral part of the protocol.
When nodes later want to retrieve content, they can find it by forwarding a query to nodes that are close to the content’s address within the overlay network. Essentially, if a node needs a specific chunk, it sends a request to the Swarm network with the address of the content. The Swarm network then forwards the request to nodes that are closer to the chunk’s address until the data is found or the request times out. This mechanism is reminiscent of a traditional distributed hash table (DHT) but includes two important and less-explored features.
One of these features is highlighted in Vitalik’s Ethereum whitepaper, where the Ethereum development core recognized the significance of Swarm’s self-sustainability. By implementing a built-in incentive system, Swarm encourages nodes to actively participate in storing and serving chunks, making the network self-sustaining. This incentive system helps maintain the availability and reliability of content within the Swarm ecosystem.
In summary, Swarm allows users to upload content to its distributed chunkstore. Chunks are assigned to nodes based on their proximity within the overlay network. When retrieving content, nodes query the network, and the requests are forwarded to nodes closest to the content’s address. Swarm’s self-sustainability is achieved through its incentive system, ensuring the active participation of nodes in storing and serving chunks.
Documents and the Swarm hash
On the API layer, Swarm offers a chunker as a key component. The chunker is responsible for breaking down any readable source, such as a file or a video camera capture device, into fixed-sized chunks. These individual chunks, known as data chunks or leaf chunks, undergo a hashing process and are then synchronized with other peers in the Swarm network.
The hashes of these data chunks are further organized into larger chunks called intermediate chunks. This process continues recursively, with 128 hashes being combined to form a new chunk. As a result, the data is represented as a Merkle tree, where the root hash of the tree serves as the address used to retrieve the uploaded file.
By employing this approach, Swarm ensures efficient storage and retrieval of data. The use of fixed-sized chunks and Merkle tree structure enables effective content distribution, verification, and synchronization across the network.
